Micro-scale (1.5 mm) sulphur isotope analysis of contemporary and early Archean pyrite Manabu Nishizawa 1,2 * , Shigenori Maruyama 2 , Tetsuro Urabe 3 , Naoto Takahata 4 and Yuji Sano 4 1 Precambrian Ecosystem Laboratory, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka 237-0061, Japan 2 Department of Earth and Planetary Science, Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8551, Japan 3 Department of Earth and Planetary Science, The University of Tokyo, Bunkyo-ku, Tokyo, Japan 4 Center for Advanced Marine Research, Ocean Research Institute, The University of Tokyo, Nakano-ku, Tokyo 164-8639, Japan Received 20 November 2009; Revised 21 February 2010; Accepted 22 February 2010 We present a method for in situ sulphur (S) isotopic analysis of significantly small areas (1.5 mm in diameter) in pyrite using secondary ion mass spectrometry (NanoSIMS) to interpret microbial sulphur metabolism in the early earth. We evaluated the precision and accuracy of S isotopic ratios obtained by this method using hydrothermal pyrite samples with homogeneous S isotopic ratios. The internal precision of the d 34 S value was 1.5% at the level of 1 sigma of standard error (named 1SE) for a single spot, while the external reproducibility was estimated to be 1.6% at the level of 1 sigma of standard deviation (named 1SD, n ¼ 25). For each separate sample, the average d 34 S value was comparable with that measured by a conventional method, and the accuracy was better than 2.3%. Consequently, the in situ method is sufficiently accurate and precise to detect the S isotopic variations of small sample of the pyrite (less than 20 mm) that occurs ubiquitously in ancient sedimentary rocks. This method was applied to measure the S isotopic distribution of pyrite within black chert fragments in early Archean sandstone. The pyrite had isotopic zoning with a 34 S-depleted core and 34 S-enriched rim, suggesting isotopic evolution of the source H 2 S from 15 to 5%. Production of H 2 S by microbial sulphate reduction (MSR) in a closed system provides a possible explanation for both the 34 S-depleted initial H 2 S and the progressive increase in the d 34 S H2S value. Although more extensive data are necessary to strengthen the explanation for the origin of the MSR, the results show that the S isotopic distribution within pyrite crystals may be a key tracer for MSR activity in the early earth. Copyright # 2010 John Wiley & Sons, Ltd. Pyrite (FeS 2 ) is a common mineral in sediments, basalt, ore deposits and meteorites. The sulphur (S) isotopic ratio of pyrite has been used to infer the origin of S and the several processes leading up to pyritization, such as physicochemical, biological and photochemical reactions. 1–3 The S isotopic ratio (d 34 S value) of sedimentary pyrite fluctuates up to 70% due to isotopic fractionation associated with biogeochemical reac- tions such as sulphate reduction, sulphide oxidation and sulphur disproportionation. 2 It is possible for the S isotopic ratio to be heterogeneous within a single crystal of sedimentary pyrite, due to secular change in the isotopic ratio of diagenetic H 2 S. Small pyrite crystals (<20 mm in diameter) are commonly observed in both modern sediments and ancient sedimentary rocks. If the S isotopic distribution of the small pyrite crystals commonly found within sedimentary rocks can be measured, information on microbial sulphur metabolism in the geological past may be obtained. Therefore, development of an analytical method for the in situ S isotopic measurement of pyrite, with lateral resolution better than few mm, is both important and desirable. Micro-scale S isotopic analyses were previously performed by ion microprobe and laser microprobe. 4–6 However, there are only a few reports on S isotopic composition with a lateral resolution of 2 mm due to technical difficulties. 7–10 Recently, a high lateral resolution secondary ion mass spectrometer (ion microprobe) with superior lateral resolution of up to 0.05 mm has been developed by Cameca. 11 Because secondary ion mass spectrometry of very small areas (NanoSIMS) is suitable for measuring large isotopic variations at the sub-micron scale, most analyses using NanoSIMS have focused on materials with artificially enriched isotope signals 12,13 and/or extra- terrestrial materials with large isotopic anomalies of up to 20 000%. 14 However, NanoSIMS has not been applied extensively to high-precision isotopic analyses 9,15–18 that are essential to investigate isotopic variations in terrestrial samples at their natural abundance levels Winterholler et al. recently evaluated the precision and accuracy of a NanoSIMS isotopic analysis of sulphate particles with diameter of 15–0.5 mm. 9 However, the precision and accuracy of S isotopic analysis of pyrite with lateral resolution better than two micrometres using RAPID COMMUNICATIONS IN MASS SPECTROMETRY Rapid Commun. Mass Spectrom. 2010; 24: 1397–1404 Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/rcm.4517 *Correspondence to: M. Nishizawa, Precambrian Ecosystem Laboratory, Japan Agency for Marine-Earth Science and Tech- nology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka 237-0061, Japan. E-mail: m_nishizawa@jamstec.go.jp Copyright # 2010 John Wiley & Sons, Ltd.